1. Volume 103, Issue 6, Pages 865-876 (December 2000)
Positive or Negative MARE-Dependent Transcriptional Regulation Is Determined by the Abundance of Small Maf Proteins 
Hozumi Motohashi, Fumiki Katsuoka, Jordan A Shavit, James Douglas Engel, Masayuki Yamamoto 
Cell 
Volume 103, Issue 6, Pages (December 2000)
DOI: /S (00)

2. Figure 1
Anemic Founder Embryos Bearing a GATA-1-directed MafK Transgene
F0 analysis of fertile oocyte injection of G1-HRD-MafK. (A and B) Representative transgenic embryos at E10.5 (TG+) are depicted alongside transgene-negative littermates (WT). (C–F) Histological examination of E10.5 F0 embryos. An embryo bearing the G1-HRD-MafK transgene (C and E; TG+) and a wild-type littermate (D and F; WT) were sectioned. (C) and (D) are cross sections stained with hematoxylin and eosin. (E) and (F) were immunostained with anti-hemoglobin antibody (light brown) and then counterstained with methyl green. Abbreviations: a, aorta; h, heart; v, cardinal vein. Scale bar corresponds to 210 μm (C and D) or 175 μm (E and F). (G) Gene expression in E9.5 F0 yolk sacs. RNA was extracted from whole yolk sacs, and semiquantitative RT-PCR was performed. Lanes 1 and 2 are samples from pale, small, very anemic transgenic embryos, whereas lane 3 was an only slightly anemic founder. Lanes 4 and 5 are wild-type control littermates. Lane 6 is water.
Cell  , DOI: ( /S (00) )

3. Figure 2
The MafK Zipper Mutant L2PM4P Does Not Bind MARE DNA nor Affect Transcription
(A) DNA binding properties of MafK and MafK L2PM4P were examined by EGMSA. MafK and MafK L2PM4P were in vitro translated and then incubated with a 32P-labeled oligonucleotide containing the MARE (NF-E2) consensus sequence (lanes 2 and 4). NF-E2 p45 was added to the reaction to examine its ability to heterodimerize (lanes 6 and 8). An arrow and an arrowhead indicate the positions of shifted bands composed of MafK homodimers or of MafK/p45 NF-E2 heterodimers, respectively. The bands were specifically competed by the addition of a 400-fold molar excess of unlabeled oligonucleotide (lanes 3 and 7). Wheat germ extract (WG) without template DNA was a negative control (lane 5).
(B) Transfection was performed in NIH3T3 cells. Relative luciferase activity is shown, and luciferase activity without addition of effector plasmids was arbitrarily set at 100%. The amount of transfected effector plasmids, pIM-MafK and pIM-MafK L2PM4P, is shown at the bottom. This represents the average of three independent experiments, and error bars indicate the standard deviation between experiments.
(C) Immunoblot analysis of nuclear extracts prepared from transfected NIH3T3 cells. NIH3T3 cells were transfected with pIM-MafK, pIM-MafK L2PM4P, or vector. Ten micrograms of each nuclear extract was applied per lane; MafK was detected by reaction with anti-p18 (MafK) antibody. Molecular size markers are on the right.
Cell  , DOI: ( /S (00) )

4. Figure 3
Small Maf Transgenic Mice Display Altered MARE Binding Activity and Inhibited Proplatelet Formation in Megakaryocytes
(A) MafK expression levels were analyzed in nuclear extracts recovered from total bone marrow cells of adult transgenic mice by immunoblotting using the anti-p18 (MafK) antibody. Line 28 bears G1-HRD-GFP, while the others coincorporated G1-HRD-MafK into the germ line.
(B–E) Proplatelet formation was analyzed in bone marrow megakaryocytes of transgenic mice. Cultured megakaryocytes from TgGFP28 (B and C) and TgMafK548 (D and E) are shown. Acetylcholinesterase activity stained dark brown (B and D). GFP-positive megakaryocytes (C and E) and proplatelet filaments (C) were observed by fluorescence microscopy.
(F) MARE binding activity in megakaryocytes isolated from MafK transgenic mice. 1.5 μg of nuclear extract from WT (lanes 1–4), TgMafK686 (lanes 5–8), or TgMafK548 (lanes 9–13) was incubated with a 32P-labeled oligonucleotide containing the NF-E2/MARE consensus sequence. The array of complexes formed (in the absence of antisera) are indicated by an arrow. WG, wheat germ extract.
(G) Total small Maf expression levels in adult transgenic bone marrow were determined using anti-small Maf antiserum (Experimental Procedures). All lines shown bear G1-HRD-MafG/GFP transgenes.
Cell  , DOI: ( /S (00) )

5. Figure 4
Proplatelet Formation and Thromboxane Synthetase Expression in Megakaryocytes of Mice Expressing Different Levels of Small Maf
(A) Small Maf expression levels in compound mutant mice were examined (see Figure 3G). Two representative high level producers (TgMafK686 and TgMafG825) are shown in both wild-type and mafG−/− backgrounds.
(B–E) Proplatelet formation was analyzed in megakaryocytes from a wild-type mouse (B), its mafG null littermate (C), a mafG−/− ::TgMafK548 mouse (D), and a mafG−/− ::TgGFP28 mouse (E). Acetylcholinesterase activity stains dark brown (B and C), and GFP-positive megakaryocytes were observed by fluorescence microscopy (D and E).
(F) Thromboxane synthetase (TXS) mRNA levels were determined by real-time quantitative RT-PCR (Experimental Procedures). The expression levels were normalized to ribosomal RNA in the same cDNA preparation. TXS mRNA was reduced in mafG null megakaryocytes compared to that in wild-type megakaryocytes (left). TXS mRNA was also reduced in megakaryocytes of two different MafK transgenic lines (548 and 549) (middle). TXS mRNA levels are fully or partially restored in mafG−/− ::TgMafK548 or mafG−/− ::TgMafK549 compound mutant mice (right). Each value represents the average of three different RNA samples, and the error bar indicates the standard deviation.
Cell  , DOI: ( /S (00) )

6. Figure 5
In Vivo Titration of Small Maf Activity Is Required for Maximum PPF
When small Maf proteins are low, such as in mafG null mutant megakaryocytes, PPF frequency is also low. In small Maf surplus, such as in megakaryocytes from MafK or MafG-overexpressing transgenic mice, PPF is also low. PPF is recovered in megakaryocytes from mafG−/− ::TgMafK/GFP or TgMafG/GFP compound mutant mice according to the expression level of the transgene.
Cell  , DOI: ( /S (00) )